Talk is cheap. Details in the execution will be the testing ground.
Trump Admin's AI action plan claims to support the notion that AI should be founded with freedom of speech and expression in mind, but then says things like "AI procured by the Federal government objectively reflects truth rather than social engineering agendas.” Who defines "truth"?
In a companion EO the administration details it's true intentions by listing potential violations for AI models related to "ideological agendas" in areas such as diversity, calling out racial bias, or valuing gender equality.
It appears that the administration's desire to prohibit top-down ideological bias in AI systems - is in fact a concrete example of pushing top-down ideological biases. https://www.wired.com/story/trump-ai-order-bias-openai-google/ #AI #AIactionplan #USGov #AISupremacy #LLMs #FreeSpeech #neutrality



